Papers by Adam Jatowt
DeAR: Dual-Stage Document Reranking with Reasoning Agents via LLM Distillation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | DeAR is an open-source framework that decouples the tasks of LLMs with holistic cross-document analysis. |
| Approach: | They propose an open-source framework that decouples relevance scoring with holistic cross-document analysis. |
| Outcome: | The proposed framework outperforms open-source frameworks in QA and open-domain QA. |
A Survey of Automatic Text Summarization Using Graph Neural Networks (2022.coling-1)
Copied to clipboard
| Challenge: | Abstractive ATS involves generating factually correct and fluent sentences. |
| Approach: | They provide an overview of the use of graph neural networks (GNNs) for automatic text summarization. |
| Outcome: | The proposed model is based on a set of graph neural networks (GNNs) that are used to generate a concise, correct and fluent summary of a given text. |
Question Difficulty Estimation for Large Language Models via Answer Plausibility Scoring (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to estimate question difficulty rely on readability formulas, retrieval-based signals, or popularity statistics. |
| Approach: | They propose a method that estimates question difficulty by computing the entropy of plausibility scores over candidate answers. |
| Outcome: | The proposed method outperforms baselines across four QA datasets and shows strong robustness across hyperparameter variations and question types. |
Exploring Union and Intersection of Visual Regions for Generating Questions, Answers, and Distractors (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing efforts to generate image-related questions, correct answers, or challenge distractors are limited. |
| Approach: | They propose to put the spotlight on different image regions to diversify QADs . they propose a framework that generates each QAD based on a recurrent multimodal encoder . |
| Outcome: | The proposed framework puts the spotlight on different image regions to diversify QADs. |
GMH: A General Multi-hop Reasoning Model for KG Completion (2021.emnlp-main)
Copied to clipboard
| Challenge: | Knowledge graphs are incomplete with many facts missing, causing performance bottlenecks in many applications. |
| Approach: | They propose a general multi-hop reasoning task that can be formulated as a search process and can be extended to long-distance reasoning scenarios. |
| Outcome: | The proposed model improves on baselines in short and long distance reasoning scenarios. |
Topic-aware Multimodal Summarization (2022.findings-aacl)
Copied to clipboard
| Challenge: | Existing work on multimodal summarization does not consider the topic of the content. |
| Approach: | They propose a topic-aware MS system which performs two tasks simultaneously: differentiating images into "on-topic" and "off-topic". |
| Outcome: | The proposed system outperforms the state-of-the-art approach by 1.7 % in ROUGE-L metric. |
Fact-Tree Reasoning for N-ary Question Answering over Knowledge Graphs (2022.findings-acl)
Copied to clipboard
| Challenge: | Current Question Answering over Knowledge Graphs (KGQA) tasks focus on binary facts, but neglect n-ary facts. |
| Approach: | They propose a new fact-tree reasoning framework that transforms the question into a fact tree and performs iterative fact reasoning on the fact tree to infer the correct answer. |
| Outcome: | The proposed framework performs iterative fact reasoning on the fact tree to infer the correct answer. |
From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations (2026.findings-acl)
Copied to clipboard
| Challenge: | DeFactoX integrates Direct Preference Optimization (DPO) with Curriculum learning to align machine-generated explanations with human reasoning. |
| Approach: | They propose a framework that integrates Direct Preference Optimization (DPO) with Curriculum learning to align machine-generated explanations with human reasoning. |
| Outcome: | The proposed framework combines Direct Preference Optimization (DPO) with Curriculum learning to align machine-generated explanations with human reasoning. |
Annotating and Analyzing Biased Sentences in News Articles using Crowdsourcing (2020.lrec-1)
Copied to clipboard
| Challenge: | a lack of publicly available news bias datasets has hindered efforts to detect subtle biases in news articles. |
| Approach: | They propose a news bias dataset which contains sentences with bias labels . they propose to use the dataset to develop and evaluate methods for detecting news bias . |
| Outcome: | The proposed dataset can be used for analyzing news bias and for developing and evaluating methods for news bias detection. |
A High-Quality Gold Standard for Citation-based Tasks (L18-1)
Copied to clipboard
| Challenge: | Citation recommendation tasks involve recommending citations within their specific contexts. |
| Approach: | They propose to use arXiv.org's citation-dependent evaluation data set to evaluate citations . their data set is characterized by the fact that it exhibits almost zero noise in its extracted content . |
| Outcome: | The proposed data set exhibits almost zero noise in extracted content and all citations are linked to their correct publications. |
How Good are LLM-based Rerankers? An Empirical Analysis of State-of-the-Art Reranking Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a systematic and comprehensive empirical evaluation of state-of-the-art reranking methods is presented. |
| Approach: | They evaluate 22 reranking methods including 40 variants across established benchmarks . primary goal is to determine whether performance disparity exists between LLM-based reranters and lightweight counterparts based on novel queries . |
| Outcome: | The proposed methods perform better on familiar queries than lightweight models, the authors show . |
Transfer Learning for Humor Detection by Twin Masked Yellow Muppets (2022.aacl-short)
Copied to clipboard
| Challenge: | Existing humor classification systems have been dealing with different forms of humor independently. |
| Approach: | They propose to combine different forms of humor to tackle different humor types by a shared-private multitask architecture using a transfer learning paradigm. |
| Outcome: | The proposed architecture shows statistically significant improvements over baselines and accounting for new state-of-the-art figures for two datasets. |
RECOR: Reasoning-focused Multi-turn Conversational Retrieval Benchmark (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks treat multi-turn conversation and reasoning-intensive retrieval separately, yet real-world information seeking requires both. |
| Approach: | They propose a framework that transforms complex queries into fact-grounded multi-turn dialogues through multi-level validation. |
| Outcome: | The proposed framework outperforms existing systems in a number of domains and can be used to improve multi-turn conversation retrieval. |
Towards the First NLP Benchmark for Ladin - an Extremely Low-Resource Language (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are limited in low-resource languages due to lack of labeled training data. |
| Approach: | They propose to use Ladin as a model for sentiment analysis and question answering by incorporating Italian data into machine translation training. |
| Outcome: | The proposed method improves on existing Italian–Ladin translation baselines. |
Exploring Hint Generation Approaches for Open-Domain Question Answering (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing automatic question answering systems rely on contextual information to provide accurate answers. |
| Approach: | They propose a context preparation approach that uses Automatic Hint Generation techniques to generate hints instead of retrieved contexts. |
| Outcome: | The proposed approach surpasses retrieval-based and generation-based methods on three QA datasets. |
BracketRank: Large Language Model Document Ranking via Reasoning-based Competitive Elimination (2026.acl-long)
Copied to clipboard
| Challenge: | Existing lists of document ranking methods lack robust performance across domains. |
| Approach: | They propose a reasoning-driven competitive elimination framework that optimises group sizes based on LLM context limits and reasoning-enhanced prompts. |
| Outcome: | The proposed method outperforms RankGPT and other state-of-the-art methods on datasets with a 77.90 NDCG@5 score and 54.66 average NDGC@10 on BEIR datasets. |
ACROSS: An Alignment-based Framework for Low-Resource Many-to-One Cross-Lingual Summarization (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies ignore data imbalance in multilingual settings and do not utilize monolingual data. |
| Approach: | They propose a cross-lingual summarization model that aligns cross-linguistic data with high-resource monolingual data via contrastive and consistency loss. |
| Outcome: | The proposed model outperforms baseline models and consistently dominates on 45 language pairs. |
CoEvo: Coevolution of LLM and Retrieval Model for Domain-Specific Information Retrieval (2025.emnlp-main)
Copied to clipboard
Ang Li, Yiquan Wu, Yinghao Hu, Lizhi Qing, Shihang Wang, Chengyuan Liu, Tao Wu, Adam Jatowt, Ming Cai, Fei Wu, Kun Kuang
| Challenge: | Recent methods to enhance queries by generating intermediary elements can degrade retrieval performance . combining LLMs and retrievers can be difficult, resulting in unreliable or irrelevant intermediaries . |
| Approach: | They propose a framework that facilitates the coevolution of large language models and retrieval models. |
| Outcome: | The proposed framework facilitates the coevolution of LLMs and retrieval models. |
Generating Questions, Answers, and Distractors for Videos: Exploring Semantic Uncertainty of Object Motions (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing video QADs are generated using video captions, incurring significant costs and resulting in bias. |
| Approach: | They propose to use temporal motion to describe video objects to generate diverse QADs focusing on different objects and interactions. |
| Outcome: | The proposed approach improves consistency and diversity of generated QADs on the NExT-QA and Perception Test benchmarks. |
CDD: A Large Scale Dataset for Legal Intelligence Research (2023.emnlp-industry)
Copied to clipboard
| Challenge: | Recent research has focused on predicting crimes, predicting outcomes of judicial debates, and extracting information from legal documents. |
| Approach: | They propose to use a large-size Court Debate Dataset to analyze court debates . they invite experienced judges to design appropriate labels for data records . |
| Outcome: | The proposed dataset includes 30,481 court cases, totaling 1,144,425 utterances. |
Listening to Patients: Detecting and Mitigating Patient Misreport in Medical Dialogue System (2025.findings-acl)
Copied to clipboard
| Challenge: | Medical Dialogue Systems (MDSs) aim to provide automated healthcare support through natural language interactions between patients and system agents. |
| Approach: | They propose a framework that detects misreports and mitigates them by generating controlled clarifying questions. |
| Outcome: | The proposed framework can detect misreports and mitigate them through generating controlled clarifying questions. |
ASRank: Zero-Shot Re-Ranking with Answer Scent for Document Retrieval (2025.findings-naacl)
Copied to clipboard
| Challenge: | Retrieval-Augmented Generation models fail to rank the most relevant documents at the top . conventional retrieval methods fail to find the most important documents . |
| Approach: | They propose a new method for scoring retrieved documents using zero-shot answer scent based on a pre-trained large language model to compute the likelihood of document-derived answers aligning with the answer scent. |
| Outcome: | The proposed method improves top-1 retrieval accuracy on NQ, TriviaQA, WebQA, ArchivalQA, HotpotQA, and Entity Questions. |
ComplexTempQA: A 100m Dataset for Complex Temporal Question Answering (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing datasets that focus on temporal knowledge are limited in size and lack comprehensive coverage of temporal information. |
| Approach: | They introduce a large-scale temporal question-answer-matching dataset . the new taxonomy categorizes questions as attributes, comparisons, and counting questions . |
| Outcome: | The proposed dataset surpasses existing benchmarks in scale and scope. |
ClaimGen-CN: A Large-scale Chinese Dataset for Legal Claim Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Currently, legal claims are not being used by non-professionals. |
| Approach: | They construct a dataset for Chinese legal claim generation task and then use it to evaluate the generated claims. |
| Outcome: | The proposed dataset is the first for the Chinese legal claim generation task and will be made publicly available. |
Temporal Validity Change Prediction (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarking tasks require models to identify temporal validity duration of a single statement . however, many data sources contain additional context, which may alter the duration of the original statement if the context is present . |
| Approach: | They propose a task benchmarking the ability of machine learning to detect context statements that induce such change. |
| Outcome: | The proposed task uses a dataset of temporal target statements and crowdsource corresponding context statements to benchmark them. |
It’s High Time: A Survey of Temporal Question Answering (2026.acl-long)
Copied to clipboard
| Challenge: | Temporal Question Answering (TQA) is a research area that focuses on answering questions involving temporal constraints or context. |
| Approach: | They present a comprehensive overview of Temporal Question Answering (TQA) this research area focuses on answering questions involving temporal constraints or context . |
| Outcome: | The proposed frameworks are compared against a range of datasets, tasks, and approaches. |
Do Language Models Have a Common Sense regarding Time? Revisiting Temporal Commonsense Reasoning in the Era of Large Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Temporal reasoning is a vital component of human communication and understanding, yet remains an underexplored area within the context of Large Language Models (LLMs). |
| Approach: | They propose to use 3 prompting strategies to evaluate 8 different LLMs across 6 datasets and 2 Code Generation LMs to perform the analysis. |
| Outcome: | The proposed models perform better on NLP tasks than the standard models on the same dataset. |
Multi-TimeLine Summarization (MTLS): Improving Timeline Summarization by Generating Multiple Summaries (2021.acl-long)
Copied to clipboard
| Challenge: | Existing work on Time-Line Summarization (TLS) has focused on improving the performance of summarization but its drawbacks are as follows: a homogeneous dataset makes it hard to generalize; output is usually a single timeline regardless of the size and complexity of the input dataset. |
| Approach: | They propose a task that generates a time-line for each story given a news article . they propose MTLS task that can be generalized to other news articles . |
| Outcome: | The proposed task can generate bet-ter results than Time-Line Summarization (TLS) the proposed task is based on previous evaluation methods. |
Chain-of-Quizzes: Pedagogy-inspired Example Selection in In-Context-Learning (2024.findings-acl)
Copied to clipboard
| Challenge: | In-context learning (ICL) is a powerful tool for enhancing large language models (LLMs) by mimicking the human learning process. |
| Approach: | They propose a Chain-of-Quizzes framework that uses LLMs to answer a quiz to sift 'good' examples, combine them iteratively with the increasing complexity, and utilize a final exam to gauge the combined example chains. |
| Outcome: | The proposed framework outperforms baseline models on diverse reasoning datasets and shows that it is scalable and can be used in future research. |
Rankify: A Comprehensive Python Toolkit for Retrieval, Re-Ranking, and Retrieval-Augmented Generation (2026.acl-demo)
Copied to clipboard
Abdelrahman Abdallah, Bhawna Piryani, Jamshid Mozafari, Andreas Herzinger, Jamie Holdcroft, Adam Jatowt
| Challenge: | Rankify unifies retrieval-augmented generation (RAG) and retrieval based question answering systems. |
| Approach: | They propose an open-source Python toolkit that unifies retrieval-augmented generation in a single modular framework. |
| Outcome: | The proposed framework unifies retrieval-augmented generation (RAG) tools in a single modular framework. |
A Study into Investigating Temporal Robustness of LLMs (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are limited in their ability to process temporal information and perform tasks requiring temporal reasoning and factual knowledge. |
| Approach: | They propose to use eight time-sensitiverobustness tests to test the model's temporal robustness for user questions in the zero-shot setting. |
| Outcome: | The proposed tests improve the temporal QA performance by up to 55%. |
Negative Sampling Techniques in Dense Retrieval: A Survey (2026.findings-eacl)
Copied to clipboard
| Challenge: | Information Retrieval (IR) is fundamental to many modern NLP applications. |
| Approach: | They propose a taxonomy that categorizes negative sampling techniques in dense IR . they analyze them with respect to trade-offs between effectiveness, computational cost, implementation difficulty . |
| Outcome: | The proposed taxonomy categorizes techniques using random, static/dynamically mined, and synthetic datasets. |
MAKED: Multi-lingual Automatic Keyword Extraction Dataset (2022.lrec-1)
Copied to clipboard
| Challenge: | a large dataset of news articles spanning 20 languages is lacking for keyword extraction. |
| Approach: | They propose a large-scale multi-lingual keyword extraction dataset for 11 of 20 languages . authors believe it will help advance the field of automatic keyword extraction . |
| Outcome: | The proposed dataset is the first for 11 of 20 languages and is based on 540K+ news articles from the BBC News network. |
DynRank: Improve Passage Retrieval with Dynamic Zero-Shot Prompting Based on Question Classification (2025.coling-main)
Copied to clipboard
Abdelrahman Elsayed Mahmoud Abdallah, Jamshid Mozafari, Bhawna Piryani, Mohammed M.Abdelgwad, Adam Jatowt
| Challenge: | Existing approaches to enhancing passage retrieval rely on static prompts and pre-defined templates. |
| Approach: | They propose a dynamic question classification framework for open-domain question-answering systems that generates contextually relevant prompts. |
| Outcome: | The proposed framework improves passage retrieval in open-domain questionanswering systems by generating contextually relevant prompts. |
Legal Judgment Prediction based on Knowledge-enhanced Multi-Task and Multi-Label Text Classification (2025.naacl-long)
Copied to clipboard
| Challenge: | Legal judgment prediction (LJP) is an essential task for legal AI, aiming at predicting judgments based on the facts of a case. |
| Approach: | They propose a knowledge-enhanced approach that incorporates 'label-level knowledge' to enhance the representation of case facts for each task and 'task-level' knowledge to improve synergy. |
| Outcome: | The proposed method is effective in comparison to state-of-the-art (SOTA) baselines. |
Focus-aware Response Generation in Inquiry Conversation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on response generation focus on relevance and fluency, rarely paying attention to the focus. |
| Approach: | They propose a Focus-aware response generation method that takes the focus into consideration and optimizes a multi-level encoder and focal decoder to generate multiple candidate responses. |
| Outcome: | The proposed method generates candidate responses that correspond to different focuses and performs better on two orthogonal inquiry conversation datasets. |
Dataset for Temporal Analysis of English-French Cognates (2020.lrec-1)
Copied to clipboard
| Challenge: | Using computational techniques to study language evolution has gained much attention . comparing two or more languages can shed light on how they co-evolve . |
| Approach: | They propose to use a dataset to investigate the similarity in evolution between languages by comparing cognates across time. |
| Outcome: | The proposed dataset is the first to use computational approaches and large data to make a cross-language diachronic analysis. |
Multilingual Epidemiological Text Classification: A Comparative Study (2020.coling-main)
Copied to clipboard
| Challenge: | a comparative study of multilingual text classification models analyzes the performance of different models based on different languages . low-resource languages are highly influenced by typology of the languages on which the models have been trained or fine-tuned but also by their size. |
| Approach: | They compare machine and deep learning models with a dataset of epidemiological news articles . they find that the performance of the models is proportionate to the training data size . |
| Outcome: | The proposed model outperforms baseline models on a multilingual text classification task . low-resource languages are highly influenced by typology of languages and their size . |
A Multi-Attention based Neural Network with External Knowledge for Story Ending Predicting Task (C18-1)
Copied to clipboard
| Challenge: | Existing studies on the topic of common sense story understanding focus on generating guesses for a missing event or concentrating on unsupervised learning. |
| Approach: | They propose to extend attention-based neural network with external knowledge resources to understand temporal stories and predict their endings. |
| Outcome: | The proposed model outperforms state-of-the-art models and external knowledge resources. |
Can We Learn Question, Answer, and Distractors All from an Image? A New Task for Multiple-choice Visual Question Answering (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing studies focus on generating QADs from image and question, but a novel task is needed to generate meaningful questions, correct answers, and challenging distractors. |
| Approach: | They propose a task to generate QADs from images and encode images together . they use contrastive learning to ensure consistency of QAD generated and tested . |
| Outcome: | Empirical evaluations on the benchmark dataset validate the performance of the proposed task. |
Detecting Temporal Ambiguity in Questions (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Ambiguous questions have different answers depending on their interpretation and can take diverse forms. |
| Approach: | They propose a manually annotated temporally ambiguous QA dataset that captures temporal ambiguity and propose different search strategies based on disambiguate versions of the questions. |
| Outcome: | The proposed approach captures temporal ambiguity and provides non-search, competitive baselines for detecting temporal and few-shot ambiguities. |